obsidian-vault

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides bash commands like find and grep to interact with a local directory. These commands are limited to the path /mnt/d/Obsidian Vault/AI Research/ and are used for file discovery and content searching.\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it reads and processes user-controlled markdown files.\n
  • Ingestion points: File reading via search workflows in SKILL.md.\n
  • Boundary markers: Absent; no delimiters are defined to separate note content from agent instructions.\n
  • Capability inventory: Uses shell commands find and grep to access the local file system.\n
  • Sanitization: Absent; no filtering or validation of note content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 01:22 AM
Security Audit — agent-trust-hub — obsidian-vault